Embed Size (px)
Citation preview
In�uence of Weight-bearing Portion on CollapseRisk of Femoral Head Necrotic AfterIntertrochanteric Curved Varus Osteotomy: A FiniteElement AnalysisYuzhu Wang ( [email protected] )
The Fifth A�liated Hospital of Southern Medical University https://orcid.org/0000-0002-0230-5452Mincong Wang
The Fifth A�liated Hospital of Southern Medical UniversityChenglong Pan
The Fifth A�liated Hospital of Southern Medical University
Research Article
Keywords: weight-bearing portion, necrotic collapse risk, �nite element analysis
Posted Date: October 20th, 2021
DOI: https://doi.org/10.21203/rs.3.rs-970784/v1
License: This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License
1
1
2
3
4
5
Title: Influence of weight-bearing portion on collapse risk of femoral head necrosis 6
after intertrochanteric curved varus osteotomy: A finite element analysis 7
Authors: Yuzhu Wang*, Mincong Wang, Chenglong Pan 8
Department of Orthopaedic Surgery, The Fifth Affiliated Hospital of Southern 9
Medical University, Guangzhou, Guangdong, China. 10
*Correspondence: Yuzhu Wang 14
Department of Orthopaedic Surgery, The Fifth Affiliated Hospital of Southern 15
Medical University, Guangzhou, Guangdong, China. 16
Tel: 86+15521015656, Email: [email protected] 17
18
19
20
21
22
23
24
25
26
2
Abstract 27
Background 28
The influence of the insufficient developmental shape of acetabulum on the collapse 29
occurrence of osteonecrosis of the femoral head (ONFH) was reported rarely after 30
intertrochanteric curved varus osteotomy (CVO). The purpose of the study was to 31
quantitatively evaluate the influence of different weight-bearing portions on collapse 32
risk of femoral head necrosis after CVO with a finite element method. 33
Methods 34
Insufficient weight-bearing portion and normal weight-bearing portion hip joint finite 35
element models of CVO (15°, 20°, 25° and 30°) for osteonecrosis of the femoral head 36
with a lesion of 60°, dividing into three types (A, B and C1) were simulated. The 37
Mises strain and collapse index were analyzed in terms of the lesion. 38
Results 39
The maximum and mean Mises strain were higher in insufficient weight-bearing 40
portion models with a positive quantitative increment of strain, especially for type C1 41
and B in no osteotomy situation. However, the collapse index was more than 1.0 in 42
type C1, even after some degree of this CVO, and in type B of insufficient 43
weight-bearing situation. 44
Conclusions 45
Progressive collapse risk was increased in the insufficient weight-bearing portion 46
situation. Thus, the decision-making of CVO for the treatment of osteonecrosis of the 47
femoral head should be different with insufficient weight-bearing portion of the 48
3
acetabulum. We recommended that the unfavorable biomechanical shape of 49
acetabulum should be treated before this CVO was performed when the necrotic type 50
was more than B. 51
Keywords: weight-bearing portion, necrotic collapse risk, finite element analysis. 52
Background 53
Strategies standing for preserving hip joint for the treatment of osteonecrosis of the 54
femoral head (ONFH) vary based on mechanisms contributed to improving the 55
biological or mechanical characteristics involving in the occurrence of non-hereditary 56
ONFH [1,2,3,4,5,6,7,8,9]. Intertrochanteric curved varus osteotomy (CVO) is one of 57
the proximal femoral osteotomy techniques [10,11,12,13,14]. It is aimed at removing 58
lesions out of the weight-bearing area, decreasing the stress subjected to the infarction 59
to prevent collapse in the early stage, and benefiting from the biomechanically 60
friendly technique widely used in Asia. 61
62
The intact ratio measured in an AP radiological view influences the clinical results 63
after CVO. Zhao. et al [15] reported that a post-operative intact ratio of 33.0% was 64
necessary if a satisfactory result was to be achieved after CVO. It has been proved 65
that the stress distribution of the infarction was beneficial from a 30° varus osteotomy 66
to reduce stress levels through much of the necrotic region [14]. It was prone to 67
collapse if located in the anterolateral weight-bearing tract [16], whereas few studies 68
have been investigated the collapse risk factors from the acetabular side (pelvic tilt, 69
center-edge) [17]. Although the necrosis suffered higher stress with less center-edge 70
4
angle investigated in our previous study [18]. From a biomechanical point, what 71
influence of different weight-bearing portions on necrosis collapse after CVO is not 72
clear. 73
74
In this study, we hypothesized: different weight-bearing portions can affect the 75
biomechanical properties of necrotic bone after CVO, but to what extent is not 76
defined. The purpose of the study was to quantitatively evaluate the influence of 77
different weight-bearing portions on collapse risk of femoral head necrosis after CVO 78
with a finite element method. 79
Methods 80
The present study was approved by the Ethical Review Committee of the Fifth 81
Affiliated Hospital of Southern Medical University and involved the examination of 82
an adult volunteer with a informed consent before quantitative computed tomography 83
(QCT) data obtained with the following parameters, slice thickness: 1.0 (mm), voltage: 84
120 (KV), current: 102.50 (mA), in which a hydroxyapatite (HA) phantom was used 85
during image acquisition to improve the estimation of bone mineral density (BMD) 86
from a volunteer (Sex: male, age: 27 years, Height: 164 cm, Bodyweight: 66kg) 87
without any musculoskeletal disease and operation history of the hip joint (Fig. 1a). 88
89
Construction of CVO hip joint 3D models. 90
The initial hip joint 3D model was constructed from QCT data by segmenting the 91
bony structure of ilium and femur using a medical image processing software 92
5
(MIMICS 22, Materialise, Belgium). The interface between the ilium and femoral 93
head was regarded as cartilage geometry, then the cartilage geometry was divided into 94
acetabular and femoral head cartilage (Fig. 1b) [19]. We used the table top plane (TTP) 95
as the referring plane to simulate CVO in this study and the methods described in (Fig. 96
1c), only the segmented femur solid model was employed to use for determining the 97
osteotomy center (not the femoral head center) and the radius measured with CAD 98
software (SolidWorks 2016, SolidWorks Corp, USA). There were four osteotomy 99
angulations (15°, 20°, 25° and 30°) determined to represent the clinical 100
operations used frequently, then segmented ilium, acetabular cartilage, femoral head 101
cartilage, and necrotic part were assembled to simulate the CVO models without 102
implant fixed. 103
104
Definition of necrotic lesion size and type. 105
CVO for the treatment of ONFH was based on the Japanese Investigation Committee 106
(JIC) classification in which the necrotic lesions were defined based on mechanical 107
character related to the weight-bearing portion of the acetabulum [20]. We used this 108
mechanical JIC classification to simulate three types (A, B, and C1), which was 109
determined in the middle coronal plane of the femoral head, with a fixed size of 110
necrotic lesion that was shaped as a conoid projecting from the femoral head center 111
with the cone angle of 60°to represent an early stage, low risk, precollapse of the 112
lesion [21]. The localization of the lesion was in the middle of the femoral head from 113
the sagittal view, just rotated in the coronal plane, lateral boundary of the three types 114
6
of lesions was located in the borderline determined by JIC classification for type A, B, 115
and C1 (Fig. 2). 116
117
Generation of CVO finite element models. 118
The weight-bearing portion varied by modifying the CE angle [22], measured in the 119
middle coronal plane. We determined two different lateral CE angles (18°, 33°) in 120
this study to represent the insufficient weight-bearing portion and normal 121
weight-bearing portion for investigating the influence of the weight-bearing portion 122
on ONFH with CVO (Fig. 3a, 3b). The insufficient weight-bearing portion finite 123
element model (CE angle:18°) was constructed using the methods from the literature 124
page [23]. The normal weight-bearing portion model (CE angle:33°) was the initial 125
configuration of the hip. Finally, a total of 30 different finite-element models 126
simulating three different types of necrosis combined with four varus osteotomies in 127
two different CE angle conditions were established. The mesh type used for all parts 128
in the model was a C3D4 tetrahedron element with the elements of 928,127. The 129
interfaces in the osteotomy were bounded as a tie. The contact of acetabular and 130
femoral head cartilage was defined as no friction. 131
Material assignment and boundary configurations 132
The bony structure of the femur was assigned with an isotropy heterogeneous material 133
property based on QCT data, briefly, the parameters used for converting HU to 134
radiographic CT density (𝜌𝑄𝐶𝑇(g/cm3) (Eq. (1)) were calculated from B-MAS200 135
phantom [24], and from 𝜌𝑄𝐶𝑇 to Ash density (𝜌𝑎𝑠ℎ(g/cm3) (Eq. (2)) [25], then the 136
7
apparent density that was calculated from Ash density with a ratio of 0.6 [26] was 137
converted to elastic modulus (Eq. (3)) [27]. 138
(1) 139 𝜌𝑎𝑠ℎ(g/cm3) = 0.877 × 𝜌𝑄𝐶𝑇 + 0.0789 (2) 140
E = 6850𝜌𝑎𝑝𝑝1.49 (3) 141
Where Eq. (1), Eq. (2), Eq (3) were used for calculating the elastic modulus of the 142
femur bone. The material properties of necrosis, cartilage, and ilium were summarized 143
in (Fig. 3). 144
145
The boundary configurations were adopted from a validated femoral head necrosis 146
finite element model [28]. A ground reaction force of 700 N for ARCO ⅡB was 147
performed, seven antagonistic muscles around hip joint were chosen and modeled as 148
axial connectors, of which the loads minimized the internal bending moment in every 149
cross-section of the femur [29], and the loading conditions were present in (Fig. 3). 150
The hip capsular ligaments were attached to the CVO models as 1D springs elements 151
[30,31]. For boundary conditions, the pubic symphysis and sacroiliac joint were fixed 152
to prevent translation and rotation. Finite element analysis for each CVO model was 153
performed using ABAQUS software (ABAQUS 2019, Dassault Systemes, France) in 154
the International Society of Biomechanics (ISB) coordinate system [32]. 155
156
The Mises strain was used and calculated (Eq. (4), 𝜀𝑒𝑞𝑣: Mises Strain,𝜀1: Max strain (LE) 157
𝜀2: Mid strain (LE), 𝜀3: Min strain (LE), ν: Poisson’s ratio) for observing the biomechanical 158
𝜌𝑄𝐶𝑇(g/cm3) =0.9863HU-2.0804
8
properties changes of necrotic bone differing from the Mises stress. The increment of 159
strain was calculated by values of Mises strain in insufficient weight-bearing portion 160
models minus the values of Mises strain in normal weight-bearing portion models. 161
The collapse index (CI) was also calculated (Eq. (5), εmax: maximum principled strain 162
at each element of the lesion, εlim: the ultimate strain (ultimate tensile strain εlim = 163
0.0073, ultimate compressive strain εlim = 0.0104)) [33] for evaluating the collapse 164
risk of necrotic bone in different weight-bearing portions along with CVO. The value 165
with 1.0 of CI was regarded as the cut-off point of collapse of necrotic bone. 166 𝜀𝑒𝑞𝑣 = 11+𝜈 √12 {(𝜀1 − 𝜀2)2 + (𝜀2 − 𝜀3)2 + (𝜀3 − 𝜀1)2} (4) 167
CI = εmax / εlim (5) 168
169
Validation of finite element models 170
The validation of FE models was performed with sensitivity studies and mesh 171
convergence tests, which was reported clearly in our previous study [18]. Since direct 172
experimental measurements of stress within the necrotic region along the osteotomy 173
model derived from the specimen is difficult to achieve. The physiologic 174
reasonableness and internal consistency must be evaluated for validation from the 175
present model, although the finite element model was simulated by adopting a 176
validated femoral head necrotic model from literature. The present results show 177
physiologically reasonable compressive stress distribution in the medial cortex, tensile 178
stresses in the lateral cortex and in the lateral greater trochanter, and osteotomy 179
interface was compressive stress distribution in the medial and tensile stresses in the 180
9
lateral site. For the check of internal consistency, preferential longitudinal 181
compressive-stress transmission through the primary trabeculation system of the 182
femoral head across a coronal head midsection layer was simulated in no necrosis 183
natural model and it was interrupted in the no osteotomy necrosis model, which was 184
consistent with the results of other established studies [14,21,28,34,35]. 185
186
Results 187
Mises strain of necrotic bone 188
The maximum and mean Mises strain of necrotic bone was higher in insufficient 189
weight-bearing models (Fig. 4). Particularly, in no osteotomy situation, the maximum 190
Mises strain was 0.031 vs 0.022 for type C1, 0.018 vs 0.012 for type B. 191
Correspondingly, the stain highlighted in the necrotic bone at the interface of healthy 192
bone, and was weakened along with CVO angle increasing (Fig. 5, 6). For type A, a 193
slight strain increase was observed for insufficient weight-bearing models compared 194
with those of types B and C1 (Fig. 4b, d). 195
Collapse index of necrotic bone 196
The collapse index was more than 1.0 in type C1 with less than 25° of the normal 197
weight-bearing models and with less than 20° of the insufficient weight-bearing 198
models. In addition to the situation of type B of no osteotomy insufficient 199
weight-bearing model (Fig. 7). The distribution of collapse index highlighted for type 200
C1 and type B in insufficient weight-bearing models at the interface with healthy 201
bone (Fig. 8, 9). 202
10
203
Discussion 204
This is the first biomechanical simulation using Mises strain as the mechanical 205
parameter to evaluate the effect of the weight-bearing portion on potential 206
deformation of necrotic bone pre and post CVO. Our main findings were: (1) the 207
Mises strain was higher in insufficient weight-bearing models and decreased along 208
with CVO increasing, respectively, which supported the hypothesis, (2) the collapse 209
index was more than 1.0 in type C1 and type B of insufficient weight-bearing 210
situation, suggesting that higher collapse risk was predicted for clinical practice. 211
212
Less weight-bearing portion of acetabulum such as developmental dysplasia of the hip 213
(DDH) can elevate the contact pressure [30]. Because of insufficient contact area from 214
the unfavorable shape of acetabulum, resulting in accelerating degenerative changes 215
of cartilage [36]. Moreover, the possibility of collapse for the lesion increased in the 216
situation of abnormal configurations of the hip [16]. Correspondingly, in this study, 217
the Mises strain was higher in insufficient weight-bearing portion models compared to 218
those normal. 219
220
Although CVO decreased the strain level positively, the increment of strain at each 221
osteotomy degree approved that the influence of less weight-bearing area on the 222
necrotic bone did not vanish. This is suggesting that high collapse risk exist although 223
the operations such as CVO were performed in the femur side only if unfavorable 224
11
condition of the hip from acetabulum is not improved. The results obtained from finite 225
element analysis was consistent with the clinical research focusing on CVO for 226
ONFH17, in which CE angle < 25° was one of the factors identified to influence 10 227
influence 10-year survival(56%) with radiographic failure compared to LCE angle ≥228
25° with radiographic failure in 10-year survival (77%). Thus, the recommendation of 229
what procedure for the treatment of ONFH is better in the case of the dysplastic 230
acetabulum, whatever, not the CVO as one-way approach. 231
232
The types of the lesion were decided in this study based on mechanical characteristics. 233
The effect of the insufficient weight-bearing portion on type C1 was more obvious 234
than type B and A. Interestingly, the max Mises strain increment was negative for type 235
C1 at 30° and type B at 15°, 20°, 25°. The reason maybe was that less load was 236
suffered to the necrotic bone after CVO in insufficient weight-bearing portion models. 237
Such phenomenon featured that local concentrated non-uniform biomechanical 238
distribution (stress, strain) was for the insufficient weight-bearing portion compared 239
to the normal with uniform regular pattern37. 240
241
The location and size of the lesion can influence the collapse risk even after the 242
persevering hip operations7. In this study, the collapse index was used for evaluating 243
the influence of weight-bearing portion on collapse risk of the lesion while 244
performing CVO. Although the collapse index was higher in insufficient 245
weight-bearing portion models of type C1, the collapse index of normal models was 246
12
more than 1.0 in less than 25° of CVO, in other words, the influence of the location of 247
lesion on collapse risk was more than the insufficient weight-bearing area for type C1. 248
Moreover, the small degree of CVO maybe not enough to eliminate the collapse risk 249
from the predicted results. 250
251
The main limitation of the study was a computational simulation of prediction, 252
although consistent with the clinical outcomes, the predicted results cannot be used as 253
a standard for clinical practice. Just providing some suggestions for better decision 254
making. Secondly, this simulation only addressed mechanical factors that would affect 255
the stain distribution of necrotic bone; however, biological factors also need to be 256
considered. 257
258
Conclusions 259
Understanding the complicated interdependence of location of necrotic lesions and the 260
configuration of the hip and pelvis is important in making decisions regarding optimal 261
treatment. This computational simulation suggested progressive collapse risk was 262
increased in the insufficient weight-bearing portion situation, and the location of 263
necrotic bone should be considered before performing persevering hip surgeries for 264
the treatment of ONFH. We recommended that the unfavorable biomechanical shape 265
of acetabulum should be treated before this CVO was performed when the necrotic 266
type was more than B. 267
Availability of data and materials 268
13
The data and materials are available from the corresponding author. 269
270
Abbreviations 271
CVO: Curved varus osteotomy 272
ONFH: Osteonecrosis of the femoral head 273
LCE: Lateral center-edge angle 274
FEA: Finite element analysis 275
QCT: Quantitative computed tomography 276
JIC: Japanese Investigation Committee 277
HU: Hounsfield unit 278
Eq: Equation 279
CAD: Computer assisted design 280
vs: Versus 281
3D: Three-dimensional 282
DDH: Developmental dysplasia of the hip 283
References 284
1 Wang T, Azeddine B, Mah W, Harvey EJ, Rosenblatt D, Séguin C. Osteonecrosis 285
of the femoral head: genetic basis. Int Orthop, 2019, 43(3): 519-30. 286
2 Hernigou P, Poignard A, Zilber S, Rouard H. Cell therapy of hip osteonecrosis 287
with autologous bone marrow grafting. Indian J Orthop, 2009, 43(1): 40-5. 288
3 Steinberg ME, Brighton CT, Steinberg DR, Tooze SE, Hayken GD. Treatment of 289
avascular necrosis of the femoral head by a combination of bone grafting, 290
decompression, and electrical stimulation. Clin Orthop Relat Res, 1984, (186): 291
137-53. 292
14
4 Korompilias AV, Lykissas MG, Beris AE, et al. Vascularised fibular graft in the 293
management of femoral head osteonecrosis: Twenty years later. J Bone J Surg 294
(Br), 2009, 91(3): 287-93. 295
5 Hernigou P, Beaujean F. Treatment of osteonecrosis with autologous bone 296
marrow grafting. Clin Orthop Reslat Res, 2002, (405): 14-23. 297
6 Aarvold A, Smith JO, Tayton ER, Jones AM, Dawson JI, Lanham S, Briscoe A, 298
Dunlop DG, Oreffo RO. A tissue engineering strategy for the treatment of 299
avascular necrosis of the femoral head. Surgeon, 2013, 11(6): 319-25. 300
7 Mont MA, Carbone JJ, Fairbank AC. Core decompression versus nonoperative 301
management for osteonecrosis of the hip. Clin Orthop Reslat Res, 1996, (324): 302
169-78. 303
8 Veillette CJ, Mehdian H, Schemitsch EH, McKee MD. Survivorship analysis and 304
radiographic outcome following tantalum rod insertion for osteonecrosis of the 305
femoral head. J Bone J Surg (A), 2006, 88(3): 48-55. 306
9 Keizer SB, Kock NB, Dijkstra PD, Taminiau AH, Nelissen RG. Treatment of 307
avascular necrosis of the hip by a non-vascularised cortical graft. J Bone J Surg, 308
2006, 88(4): 460-6. 309
10 Ito H, Tanino H, Yamanaka Y, Nakamura T, Takahashi D, Minami A, Matsuno T. 310
Long-term results of conventional varus half wedge proximal femoral osteotomy 311
for the treatment of osteonecrosis of the femoral head. J Bone J Surg (Br), 2012, 312
94(3): 308-14. 313
11 Lee YK, Park CH, Ha YC, Kim DY, Lyu SH, Koo KH. Comparison of surgical 314
parameter and results between curved varus osteotomy and rotational osteotomy 315
for osteonecrosis of the femoral head. Clin Orthop Surg, 2017, 9(2): 160-168. 316
12 Sakano S, Hasegawa Y, Torii Y, Kawasaki M, Ishiguro N. Curved 317
intertrochanteric varus osteotomy for osteonecrosis of the femoral head. J Bone J 318
Surg (Br), 2004, 86(3): 359-65. 319
13 Nozawa M, Enomoto F, Shitoto K, Matsuda K, Maezawa K, Kurosawa H. 320
Rotational Acetabular Osteotomy for Osteonecrosis with Collapse of the Femoral 321
15
Head in Yong Patients. J Bone J Surg (A), 2005, 87(3): 514-20. 322
14 Baker KJ, Brown TD, Brand RA. A finite-element analysis of the effects of 323
intertrochanteric osteotomy on stresses in femoral head osteonecrosis. Clin 324
Orthop Relat Res, 1989, (249): 183-98. 325
15 Zhao G, Yamamoto T, Ikemura S, Motomura G, Mawatari T, Nakashima Y, 326
Iwamoto Y. Radiological outcome analysis of transtrochanteric curved varus 327
osteotomy for osteonecrosis of the femoral head at a mean follow-up of 12.4 328
years. J Bone J Surg (Br), 2010, 92(6): 781-6. 329
16 Roush TF, Olson SA, Pietrobon R, Braga L, Urbaniak JR. Influence of 330
Acetabular Coverage on Hip Survival After Free Vascularized Fibular Grafting 331
for Femoral Head Osteonecrosis. J Bone J Surg (A), 2006, 88(10): 2152-8. 332
17 Okura T, Hasegawa Y, Morita D, Osawa Y, Ishiguro N. What factors predict the 333
failure of curved intertrochanteric varus osteotomy for the osteonecrosis of the 334
femoral head? Arch. Orthop. Trauma Surg, 2016, 136(12): 1647-55. 335
18 Wang Y, Yamako G, Okada T, Arakawa H, Nakamura Y, Chosa E. Biomechanical 336
effect of intertrochanteric curved varus osteotomy on stress reduction in femoral 337
head osteonecrosis: a finite element analysis. Biomechanical effect of 338
intertrochanteric curved varus osteotomy on stress reduction in femoral head 339
osteonecrosis: a finite element analysis. J Orthop Surg Res, 2021,23;16(1):465. 340
19 Henak CR, Carruth ED, Anderson AE, Harris MD, Ellis BJ, Peters CL, Weiss JA. 341
Finite element predictions of cartilage contact mechanics in hips with retroverted 342
acetabula. Osteoarthr Cartil, 2013, 21(10): 1522-9. 343
20 Sugano N, Atsumi T, Ohzono K, Kubo T, Hotokebuchi T, Takaoka K. The 2001 344
revised criteria for diagnosis, classification, and staging of idiopathic 345
osteonecrosis of the femoral head. J Orthop Sci, 2002, 7(5): 601-5. 346
21 Lee MS, Tai CL, Senan V, Shih CH, Lo SW, Chen WP. The effect of necrotic 347
lesion and rotational degree on the stress reduction in transtrochanteric rotational 348
osteotomy for femoral head osteonecrosis – a three-dimensional finite-element 349
simulation. Clin Biomech, 2006, 21(9): 969-76. 350
16
22 Wylie JD, Kapron AL, Peters CL, Aoki SK, Maak TG. Relationship Between the 351
Lateral Center-Edge Angle and 3-Dimensional Acetabular Coverage. Orthop J 352
Sport Med, 2017, 5(4): 2325967117700589. 353
23 Zhao X, Chosa E, Totoribe K, Deng G. Effect of periacetabular osteotomy for 354
acetabular dysplasia clarified by three-dimensional finite element analysis. J 355
Orthop Sci, 2010, 15(5): 632-40. 356
24 Knowles NK, Reeves JM, Ferreira LM. Quantitative Computed Tomography 357
(QCT) derived Bone Mineral Density (BMD) in finite element studies: a review 358
of the literature. J Exp Orthop, 2016, 3(1): 36. 359
25 Schileo E, Dall'ara E, Taddei F, Malandrino A, Schotkamp T, Baleani M, 360
Viceconti M. An accurate estimation of bone density improves the accuracy of 361
subject-specific finite element models. J Biomech, 2008, 41(11): 2483-91. 362
26 Ali AA, Cristofolini L, Schileo E, Hu H, Taddei F, Kim RH, Rullkoetter PJ, Laz PJ. 363
Specimen-specific modeling of hip fracture pattern and repair. J Biomech. 2014 Jan 364
22;47(2):536-43. doi:Ali AA, Cristofolini L, Schileo E, et al. Specimen-specific 365
modeling of hip fracture pattern and repair. J Biomech, 2014, 47(2): 536-43. 366
27 Morgan EF, Bayraktar HH, Keaveny TM. Trabecular bone modulus-density 367
relationships depend on anatomic site. J Biomech, 2003, 36(7): 897-904. 368
28. Zhou GQ, Pang ZH, Chen QQ, et al. Reconstruction of the biomechnical 369
transfer path of femoral head necrosis: A subject-specific finite element 370
investigation. Comput Biol Med, 2014, 52: 96-101. 371
29 Sverdlova NS, Witzel U. Principles of determination and verification of muscle 372
forces in the human musculoskeletal system: Muscle forces to minimize bending 373
stress. J Biomech, 2010, 43(3): 387-96. 374
30 Wang X, Peng J, Li D, Zhang L, Wang H, Jiang L, Chen X. Does the optimal 375
position of the acetabular fragment should be within the radiological normal 376
range for all development dysplasia of the hip? A patient-specific finite element 377
analysis. J Orthop Surg Res, 2016, 11(1): 109. 378
31 Zou Z, Chávez-Arreola A, Mandal P, Board TN, Alonso-Rasgado T. 379
17
Optimization of the position of the acetabulum in a ganz periacetabular 380
osteotomy by finite element analysis. J Orthop Res, 2013, 31(3): 472-9. 381
32 Wu G, Siegler S, Allard P, Kirtley C, Leardini A, Rosenbaum D, Whittle M, 382
D'Lima DD, Cristofolini L, Witte H, Schmid O, Stokes I. ISB recommendation 383
on definitions of joint coordinate system of various joints for the reporting of 384
human joint motion – Part Ⅰ: Ankle, hip, and spine. J Biomech, 2002, 35(4): 385
543-8. 386
33 Xie P, Deng Y, Tan J, Wang M, Yang Y, Ouyang H, Huang W. The effect of 387
rotational degree and routine activity on the risk of collapse in transtrochanteric 388
rotational osteotomy for osteonecrosis of the femoral head—a finite element 389
analysis. Med Biol Eng Comput, 2020, 58(4): 805-14. 390
34 Li TX, Huang ZQ, Li Y, Xue ZP, Sun JG, Gao HH, He HJ, Chen WH. Prediction 391
of Collapse Using Patient-Specific Finite Element Analysis of Osteonecrosis of 392
the Femoral Head. Orthop Surg, 2019, 11(5): 794-800. 393
35. Yang JW, Koo KH, Lee MC, Yang P, Noh MD, Kim SY, Kim KI, Ha YC, Joun 394
MS. Mechanics of femoral head osteonecrosis using three-dimensional finite 395
element method. Arch. Orthop Trauma Surg, 2002, 122(2): 88-92. 396
36. Vafaeian B, Zonoobi D, Mabee M, Hareendranathan AR, El-Rich M, Adeeb S, 397
Jaremko JL. Finite element analysis of mechanical behavior of human dysplasia 398
hip joints: a systematic review. Osteoarthr Cartil, 2017, 25(4): 438-47. 399
37 Nozomi K, Takehiro I, Kimio S, Komatsu A, Yoichi S. Finite Element Analysis 400
of the Efficacy of Shelf Acetabuloplasty for Acetabular Dysplasia. Int J Phys 401
Med Rehabil, 2018, 6(6): 500. 402
403
Acknowledgements 404
I would like to thank Min-cong Wang for using his valuable time to write the 405
manuscript together and providing a lot of suggestions. 406
18
407
Funding 408
This work was supported by Research Program of Traditional Chinese Medicine 409
Bureau of Guangdong Province with Grant Number 20202122. The funder had no 410
role in design of the study, the collection, analysis, and interpretation of the data, or in 411
writing the manuscript. 412
413
Author information 414
Affiliations 415
Department of Orthopaedic Surgery, The Fifth Affiliated Hospital of Southern 416
Medical University, Guangzhou, Guangdong, China. 417
Yuzhu Wang, Mincong Wang, Chenglong Pan 418
Contributions 419
YW and MW contributed to the study equally, as the first co-authors. (I) Conception 420
and design: YW; (II) Administrative support: MW; (III) Provision of study materials 421
or patients: YW; (IV) Collection and assembly of data: MW; (V) Data analysis and 422
interpretation: YW; (VI) Manuscript writing: All authors; (VII) Final approval of 423
manuscript: All authors. 424
425
Corresponding author 426
Correspondence to YW. 427
428
Ethics Declarations 429
Ethics approval and consent to participate 430
19
The study was approved by the Ethical Review Committee of Southern Medical 431
University (NO. 21-0223), and informed consent was obtained from subjects. 432
433
Consent for publication 434
Not applicable 435
436
Competing interests 437
The authors declare no conflicts of interest in association with the present study. 438
439
Figure legends 440
Fig. 1 The solid 3D model of CVO. a Parameters of CT data obtained from the 441
subject, b The hip joint model constructed from CT images, c the method of 442
CVO, MFC: medial femoral condyle, LFC: lateral femoral condyle, PF: 443
proximal femur. 444
Fig. 2 Necrotic simulation using the classification of the Japanese Investigation 445
Committee of Health and Welfare: type A, type B, type C1. 446
Fig. 3 Finite element models of CVO. a Normal weight-bearing portion model. b 447
Insufficient weight-bearing portion model. 448
Fig. 4 Maximum (increment) and mean (increment) Mises strain of necrotic bone. 449
Fig. 5 Mises strain distribution of necrotic bone in top view. 450
Fig. 6 Mises strain distribution of necrotic bone in central coronal view. 451
Fig. 7 Maximum collapse index of necrotic bone in each model. 452
Fig. 8 Collapse index distribution of necrotic bone in top view. 453
Fig. 9 Collapse index distribution of necrotic bone in central coronal view. 454
455
Figures
Figure 1
The solid 3D model of CVO. a Parameters of CT data obtained from the subject, b The hip joint modelconstructed from CT images, c the method of CVO, MFC: medial femoral condyle, LFC: lateral femoralcondyle, PF: proximal femur.
Figure 2
Necrotic simulation using the classi�cation of the Japanese Investigation Committee of Health andWelfare: type A, type B, type C1.
Figure 3
Finite element models of CVO. a Normal weight-bearing portion model. b Insu�cient weight-bearingportion model.
Figure 4
Maximum (increment) and mean (increment) Mises strain of necrotic bone.
Figure 5
Mises strain distribution of necrotic bone in top view.
Figure 6
Mises strain distribution of necrotic bone in central coronal view.
Figure 7
Maximum collapse index of necrotic bone in each model.
Figure 8
Collapse index distribution of necrotic bone in top view.
Figure 9
Collapse index distribution of necrotic bone in central coronal view.
